Home / Technology / AI & IT / Next-generation maritime inertial navigation systems addressing the navigation and localization challenge of Autonomous Underwater Vehicles (AUV), critical for commercial and military missions

Next-generation maritime inertial navigation systems addressing the navigation and localization challenge of Autonomous Underwater Vehicles (AUV), critical for commercial and military missions

Autonomous Underwater Vehicles (AUV), also known as unmanned underwater vehicle, conducts its survey mission without operator intervention. Their autonomy allows AUVs to be used for missions where a surface vehicle or manned submersible would be at risk, such as mine countermeasure (MCM),  under-ice operations or underwater survey missions such as detecting and mapping submerged wrecks, rocks, and obstructions.


When a mission is complete, the AUV will return to a pre-programmed location and the data collected can be downloaded and processed in the same way as data collected by shipboard system. For an AUV to successfully complete a typical survey mission, it must follow a path specified by the operator as closely as possible and arrive at a precise location for collection by a surface vessel. If the final position of the AUV is not accurate, the AUV may be unrecoverable.


If the AUV does not follow the path accurately during the mission, critical features may not be recorded and the position of any features recorded during the mission will be uncertain. Accurate localization and navigation is essential to ensure the accuracy of the gathered data for these applications. A distinction should be made between navigation and localization. Navigational accuracy is the precision with which the AUV guides itself from one point to another. Localization accuracy is the error in how well the AUV localizes itself within a map.


AUV navigation and localization is a challenging problem due primarily to the rapid attenuation of higher frequency signals and the unstructured nature of the undersea environment. Above water, most autonomous systems rely on radio or spread spectrum communications and global positioning. However, underwater such signals propagate only short distances and acoustic based sensors and communications perform better.


Unlike unmanned underwater vehicles (UUVs) which are usually operated remotely by an acoustic modem link, AUVs present a uniquely challenging navigational problem because they operate autonomously in a highly unstructured environment where satellite-based GPS navigation is not directly available.

Therefore autonomous aerial vehicles, AUVs must navigate using other methods when submerged.

Navigational Methods

The type of navigation system used is highly dependent on the type of operation or mission and that in many cases different systems can be combined to yield increased performance. The most important considerations are the size of the region of interest and the desired localization accuracy.

The different methods which are currently used for AUV navigation are:

  1. Magnetic compass

Magnetic compass is a relatively inexpensive sensor that can provide a heading reference by measuring the magnetic field vector. This type of compass is subject to bias in the presence of objects with a strong magnetic signature and points to the earth’s magnetic north pole. More common in marine applications, a gyrocompass measures heading using a fast spinning disc and the rotation of the earth. It is unaffected by metallic objects and points to true north. The typical accuracy is within 1 to 2 degrees for moderately priced unit.


  1. Inertial

Use a combination of accelerometers and gyroscopes (and sometimes magnetometers) to estimate a vehicle’s orientation, velocity, and gravitational forces. In this capacity, the data collected from the IMU’s sensors allow a computer to track a craft’s position, using a method known as dead reckoning. A computer continually calculates the vehicle’s current position. First, for each of the six degrees of freedom (x,y,z and θx, θy and θz), it integrates over time the sensed acceleration, together with an estimate of gravity, to calculate the current velocity. Then it integrates the velocity to calculate the current position.

Gyroscope: Measures angular rates. For underwater applications, the following two categories are widely used:

Ring Laser / Fibre Optic: Light is passed either through a series of mirrors (ring laser) or fibre optic cable in different directions. The angular rates are determined based on the phase change of the light after passing through the mirrors or fibre. Some of the  ring-laser gyroscope systems are dithered,  that makes a noise – it ‘sings’ in the water at about 1.5 kHz – this is disadvantage in covert operation like giving away the position of the submarine.

The fibre-optic gyros are silent however the performance is not quite as good as ring-laser gyros.

MEMS: An oscillating mass is suspended within a spring system.

Accelerometer: Measures the force required to accelerate a proof mass. Common designs include pendulum, MEMS, and vibrating beam among others. Accelerometer – Bias range from 0.01mg (MEMS) to 0.001mg (Pendulum)

A major disadvantage of using IMUs for navigation is that they typically suffer from accumulated error, including Abbe error. Because the guidance system is continually adding detected changes to its previously-calculated positions, any errors in measurement, however small, are accumulated from point to point. This leads to ‘drift’, or an ever-increasing difference between where the system thinks it is located, and the actual location. Drift extremely variable from 0.0001o/hr (RLG) to 60 deg/hr or more for MEMS.


Northrop Grumman unveiled a new compact next-generation maritime inertial navigation system at the Sea-Air-Space symposium.

The Sea Fiber Optic Inertial Navigation with Data Distribution (SeaFind) system features enhanced fibre-optic gyro technology, which provides it with the same level of performance as the company’s Mk 39 ring laser gyrocompass family of inertial navigation products, while being significantly smaller in size. The compact sensor measures 250×250×127 mm and weighs just 4.9 kg.


Speaking to Jane’s, Susan Lockey Hawkins, director, business, development, and strategy for Northrop Grumman’s maritime systems division, said SeaFind was developed to meet the needs of the smaller ship market. Applications include guidance systems for unmanned underwater vehicles and unmanned surface vehicles, coastal and offshore patrol vessels, and other small and medium surface vessels. “The Mk 39 is about the size of a refrigerator – so there is a huge difference in size compared to SeaFind, which means now you can have the Mk 39 calibre of inertial navigation on smaller ships. And this is fully exportable, so the market capability for this is wide open.”


SeaFind provides information for GPS-available and GPD-denied environments as well as attitude, heading, and velocity data for fire-control stabilisation and weapons initialisation. The system features a modular system architecture and comprises an inertial measurement unit (IMU) and a separate electronics unit (EU) connected via a single cable. The smaller coil size and denser IMU package allows for flexible installation where space is limited.



In order to improve performance for longer missions, DVL sonar can be used to measure the speed of the sea floor relative to the AUV. DVL emits a multibeam acoustic ping with the resulting response measured in terms of its frequency shift – i.e. Doppler shift – which translates to a velocity relative to the reflection point, with the velocity measured in three directions (two horizontal, one vertical), enabling the calculation of the vehicle’s position underwater and for the measurement of currents. This approach is most effective when AUVs are operating close to the seafloor.

Similarly, acoustic Doppler current profiler (ADCP) sonar can measure the relative speed of the local current.

DVL sonars have a limited range and can only be used when the AUV is near to the sea floor. However, when using both INS and DVL, the estimate of position is still subject to drift over time. If such a system is used to perform long missions, it must reset this navigational drift by determining its position relative to an external reference point. This can be done directly by resurfacing and using a GPS receiver but this is undesirable during deep water surveys and impossible if the surface is inaccessible due to ice.

  1. Acoustic navigation

Acoustic navigation uses acoustic transponder beacons to allow the AUV to determine its position based on measuring the time-of-flight (TOF) of signals from acoustic beacons or modems. The most common methods for AUV navigation are long baseline (LBL) which uses at least two, widely separated transponders and ultra-short baseline (USBL) which generally uses GPS-calibrated transponders on a single surface vessel.

LBL systems require the installation of at least two beacons, usually on the sea floor, which immediately return an acoustic signal sent to them by the AUV. Using knowledge of the beacon positions, the local sound speed and the time of flight of the signal, the AUV can deduce its position from the intersection of the AUV’s possible positions relative to each beacons.

USBL systems use a single beacon, usually attached to a surface ship. The current generation of USBL systems equip the beacon with an INS/GPS system to reduce the calibration requirements of the surface vessel.

Both methods have a range limited by the extent of the transponder network which is around 10km for individual LBL and 4km for USBL networks in deep water. In shallow water, the range of a USBL system can drop to less than 500m. There is no theoretical limit to the extent of a beacon network but the cost of installation and maintenance makes such an approach impractical for many missions. Both systems require calibrated and aligned beacons and corresponding calibration and programming of the AUV.

If the location of the beacons is not provided in advance, simultaneous location and mapping (SLAM) techniques can be used to generate a map of the beacon network and use it to aid navigation.

  1. Geophysical navigation

Geophysical navigation uses physical features of the AUV’s environment to produce an estimate of the location of the AUV. Geophysical navigation systems, or terrain navigation systems, use observable physical features to obtain an estimate of the AUV’s position. These can be pre-existing or purposefully deployed features.

This can be done either by supplying the AUV with an existing map of the area or by constructing such a map over the course of the AUV mission. While techniques which use local magnetic or gravitational variations have been proposed along with methods for their use, the performance of an operational system has not been published. While there is extensive evidence that similar systems are used by mammals, the unavailability of suitable commercial sensors has limited research in this area.

This must be done with sensors and processing that are capable of detecting, identifying, and classifying some environmental features.

Magnetic: It has been proposed to use magnetic field maps for localization. Although no recent publications have been found, a team at the University of Idaho has been mapping the magnetic signatures of Navy vessels.

Optical: Use of a monocular or stereo camera to capture images of the seabed and then match these images to navigate. Underwater cameras can be used as reliable, high quality sensors within a restricted range and any AUV dependent on an optical sensor for navigation would need to operate close to the sea floor.

Sonar: Sonar sensors are based on acoustic signals, however, navigation with imaging or bathymetric sonars are based on detection, identification, and classification of features in the environment.

Used to acoustically detect then identify and classify features in the environment that could be used as navigation landmarks. With bathymetric sonar features can be extracted almost directly from assembled returns. With sidescan (imaging) sonar, feature extraction is achieved through processing of imagery.

The success of any geophysical navigation method is dependent on the presence of suitable features and the ability of the system to extract useful features from the sensor data. Autonomous feature extraction from sonar data and the associated classification problem is difficult because of the typically low resolution of the sensors and amorphous shape of natural features


  1. Sonar

A sonar is a device for remotely detecting and locating objects in water using sound. Passive sonars are listening devices that record the sounds emitted by objects in water. Active sonars are devices that produce sound waves of specific, controlled frequencies, and listen for the echoes of these emitted sounds returned from remote objects in the water. Active sonars can be categorized as either imaging sonars  that produce an image of the seabed, or ranging sonars which produce bathymetric maps.

Along-track image resolution for an imaging side-scan sonar is a function of many factors such as range, sonar frequency, and water conditions, however cross-track resolution is independent of range. For example, a Klein 5000 side-scan operating at 455kHz can achieve an along track resolution of 10cm at 38m range and 61cm at the maximum 250m range and a Klein 5900 sidescan operating at 600kHz can achieve along track resolution 5cm at 10m and 20cm resolution at the maximum 100m range. In both cases nominal cross-track resolution is 3.75cm

Resolution for a bathymetry sonar is on the order of ≈ 0.4 o − 2 o along track and ≈ 5 − 10cm cross track


State Estimation methods

The current generations of AUVs are equipped with sensors which can make use of a combination of these methods during a single mission. The different sensor data obtained from each method need to be processed together throughout a mission to obtain an optimal estimate of the vehicle position. The techniques currently used for deriving an estimate of the AUV’s position from this data are:

1) Kalman filters

The KF estimates the state of a system from a sequence of uncertain observations using a predict-update cycle. First, a predictive estimate of the next state and its uncertainty is made using an existing physical model and a statistical model which describes any uncertain factors such as process noise. This prediction is then updated using an observation of the process depending on the difference between the prediction and the observation and their uncertainties. Once this updated estimate has been calculated, a new predictive estimate can be made.


If the physical system describes the motion of an AUV, the physical model can be highly nonlinear. In this case, the assumptions of the Kalman filter break down and the optimal Bayesian estimate of the process cannot be easily found. An extended Kalman filter (EKF) can be used to extend the Kalman filter to nonlinear models by using a first-order Taylor series to approximate the nonlinear processes. The partial derivatives of the physical models are taken using a Jacobian matrix and added to the linear estimates while the predictupdate cycle remains identical to the KF. The EKF has been shown to perform well for navigation with an INS and DVL provided it is updated regularly with GPS or other reference signal. However, its performance will inevitably degrade rover time if GPS is not available.


2) Particle filters

Non-parametric representation of state distribution. Instead distribution is represented by discrete particles with associated weights. Has advantage that non-Gaussian distributions and non-linear models can be incorporated.

However, they are far more computationally intensive than Kalman filters because their accuracy depends on a large number of particles being modelled. This is a major disadvantage for AUVs which are navigating using only INS since they are often smaller vehicles which have limited power and processing and particle filters have not been used to combine INS/DVL/GPS signals for AUVs

3) Simultaneous localization and mapping (SLAM) and concurrent mapping and localization (CML) algorithms

Simultaneous localization and mapping (SLAM) is the process of a robot autonomously building a map of its environment and, at the same time, localizing itself within that environment. SLAM algorithms can be either online, where only the current pose is estimated along with the map, or full where the posterior is calculated over the entire robot trajectory.


In addition, SLAM implementations can be classified as feature based, where features are extracted (detection, identification and classification) and maintained in the state space, or view based, where poses corresponding to measurements are maintained in the state space


Vision-based navigation is also an option that presents itself, despite the operating environment, “Advanced cameras coupled to high-performance computation – that is very low power – has enormous implications in the underwater environment,” Bellingham said, adding, “You don’t tend to think of AUVs as using vision systems, we tend to think of them mainly as acoustic platforms, but there are a lot of things that you want to do underwater that can be done with vision systems and we just didn’t do them because our vehicles had to be fully autonomous and none of the vision systems previously available had the capabilities to do the useful things we wanted to do. That lets you do all kinds of interesting work, such as 3D imaging.”




About Rajesh Uppal

Check Also

Autonomous Security Robots technologies trends and markets

Introduction Terrorist activities, especially those involving suicide bombers, wreak havoc, causing widespread death and destruction. …

error: Content is protected !!